Science Contrarian review: Crookes, "On Radiant Matter" (Sheffield, 22 Aug 1879)

Reviewer: Science Contrarian (Step 2). Source: one document, 23 pages of OCR text and page images, no source gap [read]. Everything below is tagged [read] (the lecture or the Step 1 files), [web] (URL given) or [memory] (confidence given). Nothing in the Investigator's findings has been dropped; the pertinent items are carried in each claim and listed again in "Investigator's findings kept". The final weighing is left to the reader.

0. Standard applied

Judge by what was done at the bench, separate "tested and failed" from "never properly tested" from "dismissed without testing", and look at how the claim was rejected. Apply the same standard to Crookes. The lecture is a demonstration lecture: no pump, gauge, coil voltage, ray velocity, focus temperature or tabulated data is given [read: deep read, "What is not in the document"]. So most claims rest on qualitative observations that others later reproduced, rather than on numbers in this document.

1. Corrections to Step 1

1. claims.md claim 1 says "at about one millionth of an atmosphere" the gas is a fourth state. The lecture ties different effects to different pressures: best phosphorescence about 1 millionth atm (0.00076 mm Hg); mechanical effects start near 0.5 mm Hg (about 660 millionths atm) and strengthen as pumping continues; the dark space reaches the glass at that pressure [read: deep read sections 2, 3, 7]. The regime is defined by "mean free path comparable to the vessel", not by one number. Claim 1 should say so.
2. claims.md claim 4 bundles three claims: (i) the stream is negatively electrified, (ii) it is made of molecules, (iii) it causes phosphorescence. They have different histories (see claim 4 below). Claim 6 likewise bundles "same in all gases" with "10^18 molecules left"; they are judged separately below.
3. I checked the contested figure myself on a 300 dpi render of PDF p.21 [read: page image p.21]. The printed denominator is 20,000,000, so the highest vacuum is 1/20,000,000 atm (5 x 10^-8 atm, about 3.8 x 10^-5 mm Hg, about 5 x 10^-3 Pa). The digest's 1/6,754,000 is an OCR/digest error, as the deep read said. Step 1 is right. Internal check: 0.01 inch / 3 miles = 5.26 x 10^-8, which matches 1/20,000,000 [calc].
4. Molecule counts. Step 1 is right that the text prints 10^24 (a 1 followed by 24 zeros, six-digit groups) and 10^18 at 1 millionth. Check: bulb volume 4/3 pi (6.75 cm)^3 = 1288 cm^3; times Stoney's 10^21 per cm^3 = 1.288 x 10^24; divided by 10^6 = 1.29 x 10^18; the filling time at 10^8 molecules a second is 4.08 x 10^8 years [calc]. The arithmetic is internally correct. One addition Step 1 did not make: Stoney's 10^21 molecules per cm^3 is about 37 times too high against the modern value, about 2.69 x 10^19 per cm^3 at 0 C and 1 atm (Loschmidt number) [memory, high]. With the modern value the bulb holds about 3.5 x 10^22 molecules, about 3.5 x 10^16 at 1 millionth, and the filling time is about 11 million years [calc]. The rhetorical point (a "vacuum" of one millionth still holds a vast number) survives.
5. facts.json "mainstream_conflict" says the lecture's observations "agree with later accounts of cathode rays". That is correct for the observations; the interpretation (molecules, a state of matter) is what later work replaced. The Investigator's lead that Crookes later wrote "What I then called 'Radiant Matter' now passes as 'Electrons'" [web: Dyer 1920 obituary, J. Chem. Soc., via the Investigator's note, https://pubs.rsc.org] I could not trace to a primary place and date; I carry it as the Investigator's find from an obituary quotation.
6. The Investigator logged a Busch 2023 quotation about photographic fogging that is not in the project's copy [read: deep read]. I could not locate its source; it stays an open lead. It may come from a different Crookes paper.
7. Plücker and the hinged cross (Wikipedia credits Plücker, 1869). Plücker died in 1868 [memory, high], so the Wikipedia date is wrong or the credit is. The lecture credits the construction to Gimingham [read]. Step 1 and the Investigator are right to flag it.

2. Claim by claim

Claim 1. Residual gas at high exhaustion is a "fourth state of matter" (Faraday's radiant matter)

a. Mathematics. There is none in the lecture, only the kinetic-theory statement that the mean free path varies inversely with the number of molecules [read: p.2]. Consistent as far as it goes. Check of the claimed regime: with the rule of thumb that the mean free path in air is about 5-7 x 10^-3 cm divided by the pressure in torr [memory, medium], at 7.6 x 10^-4 torr (one millionth atm) it is about 7-9 cm, the same size as the 13.5 cm bulb [calc]. So the lecture's description of the best-phosphorescence regime is numerically sound. b. Established results. The collisionless regime of a rarefied gas is standard physics (free-molecular flow). The observed effects are real (claims 3-5). What does not hold is "a state as far from gas as gas is from liquid": a gas at low density changes continuously; no phase transition occurs. Crookes had not identified the carriers as something other than the gas molecules (claim 4). c. Tested. By Crookes himself (the demonstrations), then by Hittorf (1869) and Goldstein (1876) before him, and by Hertz, Schuster, Perrin and Thomson after. Nobody tested "fourth state" as such; they tested what the rays were. Puluj (1880-83) rejected it, proposing charged particles torn from the electrodes ("radiant electrode matter") and, as Busch reports, calling the fourth state "transcendental" with an inclination towards spiritualism [web: https://pmc.ncbi.nlm.nih.gov/articles/PMC10311274/]. That is Puluj's characterisation; he was a rival experimenter on the same tubes. Goldstein's ether-wave school disputed the particle view [same source]. d. If true. Thermodynamics would need a separate state at low density, with no phase boundary. Nothing in mainstream physics has to change; the framing is a label. Note on reception: modern plasma sources credit Crookes's 1879 "fourth state" as the first identification of plasma; Langmuir coined "plasma" in 1928 [web: https://en.wikipedia.org/wiki/Plasma_(physics), plasmas.org, plasma-universe.com, per the Investigator]. Crookes's own definition was a neutral collisionless gas [read], not an ionised gas. The discharge in his tubes is in fact partly ionised, so the modern credit is not baseless, but it is a retrospective reading, not what the lecture says. Blavatsky read "radiant matter" as proof of finer states of matter (Theosophy Wiki, snippet, per the Investigator); the lecture's own closing sentence about the "border-land where matter and force seem to merge" [read: p.23] invited that reading. Verdict: split. As a literal state of matter composed of residual gas molecules: Historically superseded but instructive. As a label for the collisionless regime: Established (as free-molecular flow, not as a new state). The modern plasma reading: not what Crookes defined; reported here as later reception.

Claim 2. The dark space round the negative pole is the mean free path of the residual molecules

a. Consistent as a scaling statement: the dark space lengthens as pressure falls, as the mean free path does [read: p.3]. No formula is given. b. Consistent with later work only qualitatively. The cathode dark space (now named the Crookes dark space) does widen as pressure falls and fills the tube at high vacuum. Its thickness is, however, set by ionisation and the cathode fall, not by the molecular mean free path. A numerical check using the rule of thumb in claim 1: at 0.5 mm Hg, where Crookes says the dark space already reaches the glass [read: p.15], the molecular mean free path in air is about 0.01 cm [calc; memory, medium on the constant], two to three orders of magnitude below the dark space of centimetres. Even allowing that electrons have a longer free path than molecules, the identification as stated is too literal. This is my own estimate from a remembered constant, not a retrieved measurement; retagged "check failed" for a web source in the budget used. c. Tested by Crookes only in the sense of watching the space grow with pumping. Stokes, referee, recommended the Bakerian Lecture (15 Jan 1879, RR/8/151) and the sequel paper (22 Apr 1879, RR/8/152), the second with "some remarks to the author, including the suggestion of a further clarifying experiment" [web: https://makingscience.royalsociety.org/items/rr_8_152, via the Investigator's note; my own fetch of the page failed on a certificate error]. The page images of both reports are online and unread. This is the most useful unread document in the project: it records what a leading physicist found unclear in 1879. d. If true: nothing; it is the standard picture of the kinetic theory, applied to a discharge. Verdict: split. Pressure scaling of the dark space: Established. Identification with the gas mean free path: Historically superseded but instructive (the quantitative check above does not support the literal identification).

Claim 3. Straight lines normal to the surface, independent of the anode, sharp shadows, magnetic deflection like projectiles

a. The projectile analogy is qualitative and coherent: more gas slows the molecules, the magnet has longer to act, the curve increases [read: p.16-17]. No law is stated; Crookes says the "laws" would be of great interest [read: p.17]. b. Consistent with every later account of cathode rays: straight lines, shadows, emission normal to the cathode, deflection by a magnet, glass fluorescence. The observations hold in any school lab today. They were reported earlier: Hittorf (1869) had straight lines, shadows and glass fluorescence; Goldstein (1876) named "cathode rays" and found emission perpendicular to the cathode [web: Busch 2023, PMC link above]. Crookes's lecture is a public demonstration that popularised them, with new engineering (focusing cup cathode, hinged cross, "tired glass"), not a first observation. Varley (1871) had already argued for the particle view [web, via the Investigator: Notes Rec. 2006, Wikipedia]. The "tired glass" result (the exposed area gives less light afterwards [read: p.11-12]) is a real reproducible effect (damage and discolouration of glass) and nothing in the record disputes it. c. Replicated many times, by supporters and opponents alike. A modern CRT paddle-wheel tube still shows it [web: https://sciencedemonstrations.fas.harvard.edu/presentations/crt-paddle-wheel]. d. Nothing needs to change. Verdict: observations Established. The "molecular bullet" interpretation: Historically superseded but instructive (the projectiles are electrons, about 1/1800 the mass of hydrogen [web: https://en.wikipedia.org/wiki/Cathode_ray]).

Claim 4. The stream is negatively electrified; two parallel streams repel; impact causes characteristic phosphorescence

a. The repulsion argument is clean logic: parallel current carriers would attract, a stream of like charges would repel [read: p.18-19]. No deflection angle, voltage or pressure is given, so no number can be checked. b. Consistent with later work for the sign of charge: Schuster found electrostatic deflection toward a positive plate; Perrin (1895) collected the charge in a closed cylinder; Thomson (1897) measured mass-to-charge about 1/1800 that of hydrogen [web: https://en.wikipedia.org/wiki/Cathode_ray]. Mutual repulsion of like-charged beams is also standard (Coulomb force, partly reduced by space charge neutralisation). Inconsistent for "molecules": the carriers are not gas molecules. c. How it was rejected, the instructive case. Hertz (1883) built a tube with deflection plates and found no bending. It was "later determined" that his tube was insufficiently evacuated, so surface charge masked the field [web: https://en.wikipedia.org/wiki/Cathode_ray]. A negative experimental result that was real, reproducible by Hertz, and wrong in cause; Schuster repeated it at better vacuum and saw deflection. This is "tested and the test was flawed", and the flaw was the vacuum, the thing Crookes had specialised in. It shows that a well-made null result can mislead for years, a point for the reader. Note also that the Hertz/Goldstein school and Crookes disagreed on interpretation partly from national and instrument traditions [web: Busch 2023]; I give no motive beyond what that source states. Phosphorescence: Crookes's list (diamond, ruby, phenakite, spodumene, emerald, calcium sulphide; ruby line near 6,895, units unstated [read: page image p.7; units Angstrom, memory medium]) is qualitative observation, and its method was extended into "Radiant Matter Spectroscopy" (1883-87) as a rare-earth analysis tool [web: archive.org items, via the Investigator]. The ruby line near 6,900 A matches the known red emission of chromium in alumina [memory, medium]. This is the part of the programme with a lasting practical result. Crookes's own later view: "What I then called 'Radiant Matter' now passes as 'Electrons'" [web: Dyer 1920, J. Chem. Soc.; not traced to a primary date]. d. If "molecules" were true, the carriers would carry gas-specific mass; they do not. Verdict: split. Negative charge, repulsion, phosphorescence: Established. "The charged bodies are molecules of the residual gas": Historically superseded but instructive.

Claim 5. The stream carries strong mechanical force and heat (paddle wheel, radiometer vanes, platinum ring, melting at the focus)

a. No mathematics, no measurement (no speed, temperature, pressure at the focus). b. Heating: Consistent. A focused electron beam deposits energy and can melt iridio-platinum; this is the principle of electron-beam heating and melting. No temperature is given, so the claim is qualitative, but there is no tension. Force: Not consistent in mechanism. Starke (1900, Ann. Phys. 3, 101-107) varied current, pressure, metallic and non-metallic surfaces and angle of incidence, and concluded that thermal effects exceeded the momentum impulse from the electrons; Thomson (1903, Conduction of Electricity Through Gases, p.501), citing Starke, concluded the wheel moves mainly through the radiometric effect [web: Harvard page above]. Wikipedia adds that Thomson estimated electron momentum would turn the wheel about one revolution per minute [web: https://en.wikipedia.org/wiki/Crookes_tube; no page reference; I did not retrieve Thomson's calculation]. I carry both as secondary reports of the primary papers, which are not in the project (Starke 1900 is the experimental test to obtain). Contrarian point against Crookes: his own platinum-ring demonstration, vanes spun "as fast" by a red-hot wire alone as by the coil [read: p.15], is exactly the observation that favours a thermal mechanism. He presented it as showing that "a hot wire will set [radiant matter] in motion". Yet it is also the earliest-known observation in this record of charged emission from a hot wire (later thermionic emission, the Edison effect of 1880 and Richardson's law) [memory, medium; not web-checked]. So the demonstration is ambiguous in both directions: it supports heat as the cause of the spin, and it hints at a real separate phenomenon. I cannot say from the lecture which. A repeat at known pressure, with the ring heated and unbiased, then biased, would separate radiometric drive from emission. c. Reproduced: the wheel moves today [web: Harvard]. The mechanism has been tested (Starke) and the strong form, momentum transfer dominates, did not hold. d. If momentum transfer were dominant, the beam particles would have to be far more massive than electrons, in conflict with Thomson's e/m. Verdict: split. Heating at the focus: Established. Motion by direct molecular impact (strong form): Tested, not supported (Starke 1900, Thomson 1903; secondary reports only). Radiometer vanes spun by a red-hot ring: observation Established as a demonstration, mechanism Never properly tested here (the lecture does not separate the thermal and emission effects).

Claim 6. Effects the same for hydrogen, carbonic acid and air; 10^18 molecules still remain at one millionth atm

a. Counting arithmetic is correct from Stoney's input; the input itself is about 37 times high (section 1.4 above). b. "Same in all gases, only beginning at different pressures" [read: p.21]. This is what one expects if the carriers are electrons, not the gas molecules (universality of e/m, which Thomson found in 1897 across gases and cathode metals [memory, medium; not web-checked]). So Crookes's own observation contradicts his interpretation: if the stream were made of molecules, a hydrogen molecule (mass 2) and a CO2 molecule (mass 44) should deflect and behave very differently in a magnet. He observed no difference, reported qualitatively, no deflection angles. The observation counts for the later view. c. Tested only by Crookes in this lecture; the universality was confirmed and explained by Thomson. d. Nothing; it is a hint pointing toward the electron. Verdict: split. Same behaviour across gases: Established (observation) and Plausible, untested as a quantitative statement in the lecture (no data). The molecule count: Established as arithmetic from its inputs; the input is off by about 37 and the argument is unaffected.

3. How the rejection happened

• Crookes's radiant-matter position was not suppressed. He published in Phil. Trans. (Bakerian Lecture 1879, refereed favourably by Stokes), in Nature (1880, vols. 22, 101-104, 125-128), and carried the method to 1887 [web: archive.org, royalsociety.org, via the Investigator]. No funding or publication block turned up. The "absence of evidence" test in prompt.txt finds nothing anomalous: the record continues. The one record that is physically hard to reach is the notebook set MS/0410 (Science Museum), described as "register slight radiation and currently unavailable" [web: https://collection.sciencemuseumgroup.org.uk, via the Investigator].
• His interpretation was contested by named rivals: Goldstein (ether waves), Hertz (null deflection result, flawed by poor vacuum), Puluj (electrode matter). The dispute was settled in his favour for the particle view and against him for the molecular identification, by Thomson 1897. That is a test result (e/m), not a consensus decree.
• Reminder against over-praising him: he was first to popularise, not to discover, most of the effects in claim 3; he asserted "negatively electrified molecules" without a measurement of charge or mass; and he gave no pump, gauge or voltage data, which makes the lecture hard to replicate quantitatively.

4. Strongest honest case for Crookes

1. Every demonstration is real. Straight lines, shadows, magnetic deflection, repulsion, fluorescence, heating: all are reproduced today.
2. He was right that the carriers carry negative charge, were repelled by like charge, and could be steered by a magnet like a projectile, years before the corpuscle was measured.
3. His observation that the effects do not depend on the gas anticipated the universality of the electron.
4. His vacuum work (with Gimingham's multiple Sprengel pump, 1/20,000,000 atm [read; web: https://en.wikipedia.org/wiki/Crookes_tube, IET archives blog]) made the effects visible; the failure of Hertz's 1883 test came from a vacuum that was not good enough.
5. The phosphorescence programme grew into a real analytic method for rare earths.

5. Where Crookes is wrong or unsupported

1. "Fourth state": a label with no phase boundary.
2. "The carriers are molecules of the gas" (superseded).
3. Dark space equals mean free path (numerically too literal at 0.5 mm Hg).
4. Strong mechanical impact: the wheel mostly turns through the radiometric effect.
5. A molecule count about 37 times too high (Stoney's figure).
6. No numbers for most demonstrations; one cannot repeat them from the lecture alone.

6. Investigator's findings kept (all, whatever the verdict)

Own work: Bakerian Lecture 1879 (DOI 10.1098/rstl.1879.0065) and "Contributions to molecular physics in high vacua" (Phil. Trans. 1879 Part II; Nature 22, 1880); "On Electrical Insulation in High Vacua" (1878); spectroscopy series 1881-87; Science Museum notebooks MS/0410 (July 1875 - July 1883) and surviving tubes incl. the ruby tube; Crookes's later "radiant matter now passes as electrons". Support by others: Stokes referee reports RR/8/151-152; C. H. Gimingham's Sprengel pump (IET SC MSS 285); Harvard paddle-wheel tube; Schuster, Perrin, Thomson (secondary reports). Related work: Hittorf 1869; Goldstein 1876; Puluj 1883 (English translation on archive.org, puluj-4th-state-of-matter); Starke 1900 and Thomson 1903; Varley 1871; plasma sources; Theosophy. Historical record: histories by Woodruff 1966, DeKosky 1983, Gay 1996/2003/2008; unresolved items: Busch's fogging quotation, the Plücker attribution, Gimingham's dates (1869-1881 vs 1870-1882). Not reached by the Investigator (time limit) and blocked sites (Papers Past, Chronicling America, Theosophy Wiki, Sage Journals) remain to be searched.

7. Verdict table

8. What would settle the open points

• Read Stokes RR/8/151 and RR/8/152 (page images online) for what he thought unclear and what "further clarifying experiment" he proposed.
• Starke 1900 (primary German text) for the thermal versus impulse data; Thomson 1903 p.501 for the one-revolution-per-minute estimate.
• A new run: hot platinum ring unbiased vs biased at known pressure to separate radiometric drive from emission.
• The Bakerian Lecture and "Contributions" papers for the pressures and the "laws of magnetic rotation" the lecture lacks.
• Web checks used: 1 Wikipedia (Cathode ray), 1 failed (Royal Society, certificate error). Still [memory] only: Loschmidt number (high), mean-free-path constant (medium), thermionic emission link (medium), Thomson's gas-independence (medium).